Agricultural machine turning track identification method, device and equipment and storage medium

Abstract

The application provides a kind of agricultural machinery turning track identification method, device, equipment and storage medium, it is related to computer technical field.The method provided by the present application: by obtaining the historical operation track data when preset agricultural machine executes preset historical operation;The historical operation track data is preprocessed, and preprocessing track data is obtained;According to the preset turning identification algorithm, the preprocessing track data is identified, and the target turning track data corresponding to the preset agricultural machine when executing the preset historical operation is obtained.By the embodiment provided by the present application, the historical operation track data of agricultural machine can be identified by turning identification algorithm to obtain the turning track data of agricultural machine, to further analyze and study the turning track data of agricultural machine, to improve the work efficiency when agricultural machine executes operation.

Description

Technical Field

[0001] This invention relates to the field of computer technology, and in particular to a method, apparatus, device, and storage medium for recognizing the turning trajectory of agricultural machinery. Background Technology

[0002] With the development of the times, agricultural machinery information technology and equipment have enabled the positioning, tracking, and remote monitoring of agricultural machinery operations. Data collection follows agricultural machinery wherever it goes. Promoting the integration of agricultural machinery and big data is imperative. We must establish a big data mindset for agricultural machinery, discover the patterns behind agricultural machinery operation data through actual statistical analysis, and use data management to promote scientific decision-making, guide production, and ultimately promote the efficient, orderly, and scientific development of agricultural machinery and agriculture.

[0003] With the continuous influx of massive amounts of agricultural machinery operation data and geographic location data, existing agricultural machinery remote monitoring systems can only achieve remote storage, display, and simple analysis of the data. In fact, these vast spatial trajectory points of agricultural machinery reflect different operational behaviors, such as parking in hangars, road transfers, field operations, and field turns. Simultaneously, different operational behaviors of agricultural machinery also result in varying densities of spatial movement trajectory points, meaning high-density, low-density, and medium-density operational trajectory points will appear on the map. These two factors complement each other. Combined with other parameters, important operational indicators such as the total shift time occupied by different agricultural machinery behaviors can be calculated. Therefore, in-depth mining of agricultural machinery spatial trajectory data to classify different operational behaviors of agricultural machinery, and to meet the needs of agricultural machinery cooperatives and agricultural machinery management bureaus for refined management and statistical analysis of agricultural machinery, is an urgent task.

[0004] Among them, agricultural machinery trajectory recognition is a very important research topic. According to current data, the proportion of time that agricultural machinery spends turning around is too high, which greatly reduces the efficiency of effective operation of agricultural machinery, increases costs, and causes problems of reduced production and output. Summary of the Invention

[0005] This invention aims to at least solve one of the technical problems existing in the prior art. To this end, this invention provides a method, apparatus, device, and storage medium for identifying the turning trajectory of agricultural machinery. Based on the historical operating trajectory data of the agricultural machinery, a turning trajectory identification algorithm is used to identify the turning trajectory data of the agricultural machinery, which can then be further analyzed and studied to improve the working efficiency of the agricultural machinery during operations.

[0006] In a first aspect, the present invention provides a method for recognizing the turning trajectory of agricultural machinery, comprising:

[0007] Obtain historical operation trajectory data of agricultural machinery when it performs preset historical operations;

[0008] The historical operation trajectory data is preprocessed to obtain preprocessed trajectory data;

[0009] The preprocessed trajectory data is processed by a preset U-turn recognition algorithm to obtain the target U-turn trajectory data corresponding to the preset agricultural machinery when performing the preset historical operation.

[0010] Secondly, the present invention also provides a device for recognizing the turning trajectory of agricultural machinery, comprising:

[0011] The acquisition module is used to acquire historical operation trajectory data of agricultural machinery when it performs preset historical operations;

[0012] The preprocessing module is used to preprocess the historical operation trajectory data to obtain preprocessed trajectory data;

[0013] The identification module is used to identify and process the preprocessed trajectory data according to a preset U-turn identification algorithm to obtain the target U-turn trajectory data corresponding to the preset agricultural machinery when performing the preset historical operation.

[0014] Thirdly, the present invention also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the program to implement the steps of the agricultural machinery turning trajectory recognition method described above.

[0015] Fourthly, the present invention also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the steps of the agricultural machinery turning trajectory recognition method described above.

[0016] Fifthly, the present invention also provides a computer program product, including a computer program that, when executed by a processor, implements the steps of any of the above-described methods for recognizing the turning trajectory of agricultural machinery.

[0017] This invention provides a method, apparatus, device, and storage medium for identifying the turning trajectory of agricultural machinery. The method involves acquiring historical operation trajectory data of a preset agricultural machinery performing a preset historical operation; preprocessing the historical operation trajectory data to obtain preprocessed trajectory data; and then using a preset turning-around identification algorithm to identify the preprocessed trajectory data, thereby obtaining the target turning-around trajectory data corresponding to the preset agricultural machinery performing the preset historical operation. This allows for the identification of agricultural machinery turning-around trajectory data based on its historical operation trajectory data, enabling further analysis and research of the turning-around trajectory data to improve the working efficiency of agricultural machinery during operations. Attached Figure Description

[0018] To more clearly illustrate the technical solutions in this invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this invention. For those skilled in the art, other drawings can be obtained from these drawings without creative effort.

[0019] Figure 1 This is one of the flowcharts illustrating a method for identifying the turning trajectory of agricultural machinery provided by the present invention;

[0020] Figure 2 This is the present invention. Figure 1 One of the flowcharts for step S100;

[0021] Figure 3 This is the present invention. Figure 1 One of the flowcharts for step S200;

[0022] Figure 4 This is one of the schematic diagrams of labeled farmland corresponding to the labeled operation trajectory data provided by this invention;

[0023] Figure 5 This is the present invention. Figure 3 One of the flowcharts for step S210;

[0024] Figure 6 This is the present invention. Figure 3 One of the flowcharts for step S230;

[0025] Figure 7 This is one of the schematic diagrams of trajectory point data corresponding to the agricultural machinery driving direction strategy provided by the present invention;

[0026] Figure 8 This is one of the schematic diagrams of trajectory point data corresponding to the agricultural machinery driving speed strategy provided by the present invention;

[0027] Figure 9 This is one of the schematic diagrams of trajectory point data corresponding to the agricultural machinery travel distance strategy provided by this invention;

[0028] Figure 10 The corresponding invention provided Figure 9 One of the driving distance density distribution maps;

[0029] Figure 11 This is one of the correction diagrams provided by the present invention for correction based on the driving distance strategy;

[0030] Figure 12 This is the present invention. Figure 3 One of the flowcharts for step S240;

[0031] Figure 13This is the present invention. Figure 1 One of the flowcharts for step S300;

[0032] Figure 14 This is a schematic diagram of the structure of an agricultural machinery turning trajectory recognition device provided by the present invention;

[0033] Figure 15 This is a schematic diagram of the structure of the electronic device provided by the present invention. Detailed Implementation

[0034] To make the objectives, technical solutions, and advantages of this invention clearer, the technical solutions of this invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of this invention. All other embodiments obtained by those skilled in the art based on the embodiments of this invention without creative effort are within the scope of protection of this invention.

[0035] In related technologies, agricultural machinery trajectory recognition is a very important research topic. According to current data, the proportion of time that agricultural machinery spends turning around is too high, which greatly reduces the efficiency of effective operation of agricultural machinery, increases costs, and causes problems of reduced production and output.

[0036] Based on this, the present invention proposes a method, device, equipment and storage medium for identifying agricultural machinery turning trajectory, which can identify agricultural machinery turning trajectory data through a turning recognition algorithm based on the historical operation trajectory data of agricultural machinery, so as to further analyze and study the agricultural machinery turning trajectory data and improve the working efficiency of agricultural machinery when performing operations.

[0037] The specific implementation is illustrated in the following embodiments. First, a method for recognizing the turning trajectory of agricultural machinery in an embodiment of the present invention is described.

[0038] like Figure 1 As shown, it is a schematic diagram of the implementation process of a method for identifying the turning trajectory of agricultural machinery provided by an embodiment of the present invention. The method for identifying the turning trajectory of agricultural machinery may include, but is not limited to, steps S100 to S300.

[0039] S100, acquire historical operation trajectory data of agricultural machinery when performing preset historical operations;

[0040] S200, preprocess the historical operation trajectory data to obtain preprocessed trajectory data;

[0041] S300, the preprocessed trajectory data is identified and processed according to a preset U-turn recognition algorithm to obtain the target U-turn trajectory data corresponding to the preset agricultural machinery when performing the preset historical operation.

[0042] In step S100 of some embodiments, historical operation trajectory data of a preset agricultural machine performing a preset historical operation is obtained. It can be understood that the specific execution steps may be as follows: first, the agricultural machine operation attribute information corresponding to the preset agricultural machine is determined according to preset demand conditions; then, based on a preset system interface and the agricultural machine operation attribute information, the historical operation trajectory data of the preset agricultural machine performing the preset historical operation is obtained from a preset agricultural machine operation data storage system.

[0043] In step S200 of some embodiments, the historical operation trajectory data is preprocessed to obtain preprocessed trajectory data. It can be understood that after obtaining the historical operation trajectory data of a preset agricultural machine performing a preset historical operation in step S100, the specific execution steps of step S200 can be as follows: First, the historical operation trajectory data corresponding to each preset agricultural machine is cleaned to obtain standard operation trajectory data corresponding to each preset agricultural machine; the standard operation trajectory data is labeled to obtain labeled operation trajectory data corresponding to each preset agricultural machine; then, the labeled operation trajectory data is corrected according to a preset parameter correction strategy to obtain corrected trajectory data; and finally, the initial turning trajectory data is determined based on the corrected trajectory data and the labeled operation trajectory data.

[0044] Furthermore, the preprocessed trajectory data includes at least the initial U-turn trajectory data.

[0045] In step S300 of some embodiments, the preprocessed trajectory data is identified and processed according to a preset U-turn recognition algorithm to obtain the target U-turn trajectory data corresponding to the preset agricultural machinery performing the preset historical operation. It can be understood that after performing step S200 to preprocess the historical operation trajectory data and obtain the preprocessed trajectory data, the specific execution steps of step S300 can be as follows: first, the initial U-turn trajectory data is processed using the ten-fold cross-validation method to obtain target parameters; then, the initial U-turn trajectory data is verified using the random forest algorithm and the target parameters to obtain a verification result; if the verification value corresponding to the verification result exceeds a preset standard threshold, the initial U-turn trajectory data is determined to be the target U-turn trajectory data, and the obtained target U-turn trajectory data can be used for further analysis and research.

[0046] In some embodiments, reference Figure 2 As shown, step S100 may also include, but is not limited to, steps S110 to S120.

[0047] S110, determine the agricultural machinery operation attribute information corresponding to the preset agricultural machinery according to the preset demand conditions;

[0048] S120, based on the preset system interface and the agricultural machinery operation attribute information, obtain the historical operation trajectory data of the preset agricultural machinery when performing the preset historical operation from the preset agricultural machinery operation data storage system.

[0049] In step S110 of some embodiments, the agricultural machinery operation attribute information corresponding to the preset agricultural machinery is determined according to preset demand conditions.

[0050] It is understood that in some embodiments of the present invention, the preset agricultural machinery can be multiple National III tractors. According to preset requirements, the unique agricultural machinery number of each agricultural machinery and the preset time period are obtained, thereby obtaining the agricultural machinery operation attribute information.

[0051] Furthermore, the agricultural machinery operation attribute information should include at least: agricultural machinery number information, geographical location information, and time period information.

[0052] In step S120 of some embodiments, the historical operation trajectory data of the preset agricultural machinery when performing the preset historical operation is obtained from the preset agricultural machinery operation data storage system according to the preset system interface and the agricultural machinery operation attribute information.

[0053] It is understandable that the preset system interface can be the API interface of the agricultural machinery operation data storage system. Based on the API interface and the agricultural machinery operation attribute information, the historical operation trajectory data corresponding to the preset agricultural machinery performing preset historical operations can be obtained from the agricultural machinery operation data storage system.

[0054] Furthermore, the historical operation trajectory data actually refers to the trajectory point data generated by each agricultural machine when performing preset historical operations on a single day. The trajectory point data corresponding to each agricultural machine is then combined into a trajectory dataset, which is the historical operation trajectory data.

[0055] In some embodiments of the present invention, the test data corresponding to the historical operation trajectory data can be: agricultural machinery operation data of Henan Province. This dataset is widely distributed and comes from Anyang City in northern Henan, Xuchang City in central Henan, and Xinyang City in southern Henan. The trajectory data and other working condition data are uploaded every 2 seconds, and the time period information is mainly concentrated in June and July 2021. The data of June 3rd is randomly selected from the 30 days of data. Ten agricultural machines are selected from each of Anyang City, Xuchang City and Xinyang City, for a total of 30 data points, making the data more representative.

[0056] In some embodiments, reference Figure 3 As shown, step S200 may also include, but is not limited to, steps S210 to S240.

[0057] S210, the historical operation trajectory data corresponding to each preset agricultural machine is cleaned to obtain standard operation trajectory data corresponding to each preset agricultural machine;

[0058] S220, the standard operation trajectory data is annotated to obtain an annotated operation trajectory data corresponding to each of the preset agricultural machines;

[0059] S230, The labeled operation trajectory data is corrected according to the preset parameter correction strategy to obtain corrected trajectory data;

[0060] S240, the initial turning trajectory data is determined based on the corrected trajectory data and the labeled operation trajectory data.

[0061] In step S210 of some embodiments, the historical operation trajectory data corresponding to each preset agricultural machine is cleaned to obtain standard operation trajectory data corresponding to each preset agricultural machine. It can be understood that the specific execution steps may be as follows: first, based on a preset abnormal trajectory filtering strategy, the historical operation trajectory data corresponding to each preset agricultural machine is subjected to abnormal filtering processing to obtain abnormal trajectory data corresponding to each preset agricultural machine; then, the abnormal trajectory data is removed from the historical operation trajectory data to obtain the standard operation trajectory data corresponding to each preset agricultural machine.

[0062] In step S220 of some embodiments, the standard operation trajectory data is annotated to obtain annotated operation trajectory data corresponding to each preset agricultural machine. It can be understood that after step S210 is performed to clean the historical operation trajectory data corresponding to each preset agricultural machine to obtain standard operation trajectory data corresponding to each preset agricultural machine, the cleaned standard operation trajectory data is annotated according to a preset annotation platform. First, the standard operation trajectory data obtained in step S210 is input to the annotation platform, and then the road portion is circled and annotated. During the circle selection process, labels are categorized, for example, the circled portion is labeled as 1, and the unselected portion is labeled as 0. After annotation, the trajectory of the preset agricultural machine performing the preset historical operation is output through the annotation platform. Thus, referring to... Figure 4 As shown, multiple labeled farmlands can be obtained, and each labeled farmland contains the labeled operation trajectory data.

[0063] In step S230 of some embodiments, the labeled operation trajectory data is corrected according to a preset parameter correction strategy to obtain corrected trajectory data. It can be understood that after completing step S220 to label the standard operation trajectory data and obtain labeled operation trajectory data corresponding to each preset agricultural machine, step S230 may specifically involve: calculating and processing multiple operation trajectory point data based on the geographical location information to obtain parameter value data corresponding to each operation trajectory point data; determining the correction interval range for correcting the multiple operation trajectory point data according to the parameter correction strategy and the parameter value data; performing clustering processing on the operation trajectory point data according to a preset density clustering algorithm and set parameters on a preset development platform to obtain clustering results; filtering out outlier data whose values ​​fall within the correction interval range based on the clustering results; and correcting the outlier data according to the parameter correction strategy to obtain the corrected trajectory data.

[0064] In step S240 of some embodiments, the initial U-turn trajectory data is determined based on the corrected trajectory data and the labeled operation trajectory data. It can be understood that after step S230, which corrects the labeled operation trajectory data according to a preset parameter correction strategy to obtain the corrected trajectory data, the specific execution steps of step S240 may involve comparing the parameter value data corresponding to the operation trajectory point data with the correction interval range. If the parameter value data falls within the correction interval range, the operation trajectory point data corresponding to the parameter value data is selected, and the selected operation trajectory point data is used as interval trajectory point data. The corrected trajectory data is then removed from the interval trajectory point data to obtain the initial U-turn trajectory data.

[0065] In some embodiments, reference Figure 5 As shown, step S210 may also include, but is not limited to, steps S211 to S212.

[0066] S211, based on a preset abnormal trajectory filtering strategy, perform abnormal filtering processing on the historical operation trajectory data corresponding to each preset agricultural machine to obtain abnormal trajectory data corresponding to each preset agricultural machine.

[0067] S212, Remove the abnormal trajectory data from the historical operation trajectory data to obtain the standard operation trajectory data corresponding to each preset agricultural machine.

[0068] In step S211 of some embodiments, based on a preset abnormal trajectory filtering strategy, the historical operation trajectory data corresponding to each preset agricultural machine is subjected to abnormal filtering processing to obtain abnormal trajectory data corresponding to each preset agricultural machine. It can be understood that, according to the abnormal trajectory filtering strategy, the historical operation trajectory data of each preset agricultural machine is subjected to abnormal filtering processing to obtain abnormal trajectory data corresponding to each preset agricultural machine.

[0069] Furthermore, the abnormal trajectory screening strategy includes, but is not limited to: resampling anomaly type, stationary trajectory anomaly type, static drift anomaly type, and latitude and longitude anomaly type.

[0070] Furthermore, resampling anomaly types: clear out trajectory points with a time interval of 0 seconds between two trajectory points, and retain the first trajectory point; clear out consecutive trajectory points with the same latitude and longitude, the same speed and not 0, and retain the first trajectory point.

[0071] Anomaly type of stationary trajectory: Remove trajectory points with the same latitude and longitude and a speed of 0, and keep the first trajectory point.

[0072] Static drift anomaly type: Clean up trajectory points with different latitude and longitude, zero speed and continuous trajectory, and retain the first trajectory point.

[0073] Latitude and longitude anomaly type: Since the data collection area covers multiple provinces and cities in China, if the latitude and longitude range exceeds the scope of China, then the collected points are considered to have latitude and longitude anomalies and need to be cleaned up. The cleaning method is to delete them directly.

[0074] In step S212 of some embodiments, abnormal trajectory data is removed from the historical operation trajectory data to obtain the standard operation trajectory data corresponding to each preset agricultural machine. It is understood that after performing step S211, based on a preset abnormal trajectory filtering strategy, to perform abnormal filtering processing on the historical operation trajectory data corresponding to each preset agricultural machine and obtain abnormal trajectory data corresponding to each preset agricultural machine, the abnormal trajectory data obtained in step S211 is removed from the historical operation trajectory data to obtain the standard operation trajectory data corresponding to each preset agricultural machine.

[0075] In some embodiments, reference Figure 6 As shown, step S230 may also include, but is not limited to, steps S231 to S235.

[0076] S231, Based on the geographical location information, the multiple work trajectory point data are calculated and processed respectively to obtain the parameter value data corresponding to each work trajectory point data;

[0077] S232, Based on the parameter correction strategy and the parameter value data, determine the correction interval range for correcting the multiple operation trajectory point data;

[0078] S233, according to the preset density clustering algorithm and the setting parameters set in the preset development platform, the operation trajectory point data is clustered to obtain the clustering result;

[0079] S234, Based on the clustering results, filter out the outlier data of the operation trajectory point data within the correction interval;

[0080] S235, the abnormal point data is corrected according to the parameter correction strategy to obtain the corrected trajectory data.

[0081] In step S231 of some embodiments, based on the geographical location information, the multiple work trajectory point data are calculated and processed separately to obtain parameter value data corresponding to each work trajectory point data. It can be understood that, based on the geographical location information, the multiple work trajectory point data are calculated and processed separately to obtain parameter value data corresponding to each work trajectory point.

[0082] Furthermore, the geographic location information is obtained in real time through preset sensors by sensing the preset agricultural machinery performing preset historical operations.

[0083] Furthermore, the operation trajectory point data is the data corresponding to the trajectory behavior generated when the agricultural machinery performs preset historical operations.

[0084] In some embodiments, the parameter values ​​corresponding to the driving speed, driving direction, and driving distance of the agricultural machinery when performing preset historical operations can be calculated based on geographical location information.

[0085] In step S232 of some embodiments, a correction range for correcting the multiple work trajectory point data is determined based on the parameter correction strategy and the parameter value data. It is understood that after performing step S231, which calculates and processes the multiple work trajectory point data based on the geographical location information to obtain the parameter value data corresponding to each work trajectory point data, the correction range for correcting the multiple work trajectory point data is determined based on the parameter correction strategy and the parameter value data obtained in step S231.

[0086] Furthermore, the parameter correction strategies include at least: driving speed correction strategy, driving direction correction strategy, and driving distance correction strategy.

[0087] In some embodiments, when the parameter correction strategy is a driving direction correction strategy, the result diagram of the driving direction interval determination is displayed as follows: Figure 7As shown. The driving direction of the agricultural machinery is divided into intervals of 10°, which can be distributed as ([0°, 10°], [10°, 20°]...[350°, 360°]). For example, 6° should be divided into [0°, 10°]. After dividing the driving direction of each trajectory point in the farmland, the number of directions with data distribution and the number of points in each direction are counted. The number of trajectory points in each driving direction is ranked to obtain an ordered array of direction intervals. For example, the number of points distributed in the [170°, 180°] interval is the largest, and the number of points in the [350°, 360°] interval is the second largest. According to the characteristics of agricultural machinery operation, the difference between trajectory points of one ridge and another is 180°. It can be determined that the first and second ranked intervals of the driving direction interval should be the driving direction distribution of the ridges in this field. In addition, if the second ranked interval is not the [350°, 360°] direction interval, we can look for the lower ranked ones until we find complementary direction intervals in the higher ranked direction intervals to form a direction interval pair.

[0088] In step S233 of some embodiments, the work trajectory point data is clustered according to a preset density clustering algorithm and the setting parameters set on a preset development platform to obtain clustering results. It can be understood that after executing step S232, which determines the correction interval range for correcting multiple work trajectory point data based on the parameter correction strategy and the parameter value data, the work trajectory point data is clustered according to the preset density clustering algorithm and the setting parameters set on the preset development platform to obtain clustering results.

[0089] Furthermore, the default development platform can be PyCharm.

[0090] In some embodiments, the parameters can be set as follows: the initial value of minPts is [25, 35], and the value of eps is [6, 15], where eps is the scan radius and minPts is the minimum number of points.

[0091] Density-Based Spatial Clustering of Applications with Noise (DBSCAN) is a representative density-based clustering algorithm. Unlike partitioning and hierarchical clustering methods, it defines a cluster as the largest set of density-connected points. It can divide regions with sufficiently high density into clusters and can discover clusters of arbitrary shapes in noisy spatial databases.

[0092] DBScan requires two parameters: scanning radius (eps) and the minimum number of included points (minPts). Select an unvisited point arbitrarily to start, and find all neighboring points whose distance from this point is within eps (including eps).

[0093] If the number of neighboring points is ≥ minPts, the current point and its neighboring points form a cluster, and the starting point is marked as visited. Then, the process is performed recursively to process all points in the cluster that have not been marked as visited in the same way, so as to expand the cluster.

[0094] If the number of neighboring points is < minPts, the point is temporarily marked as a noise point.

[0095] If the cluster is sufficiently expanded, that is, all points in the cluster are marked as visited, then the same algorithm is used to process the unvisited points.

[0096] In step S234 of some embodiments, based on the clustering result, outlier data whose values in the working track point data fall within the correction interval is screened out. It can be understood that after step S233 is executed, which performs clustering processing on the working track point data according to a preset density clustering algorithm and setting parameters set on a preset development platform to obtain a clustering result, the outlier data whose values in the working track point data fall within the correction interval is screened out based on the clustering result.

[0097] In some embodiments, when the parameter correction strategy is a driving speed correction strategy. It is preset that the driving speed of the agricultural machinery is relatively close to constant speed driving when traveling within a strip of a farmland, while when turning around at the headland, the speed changes and presents a non-constant speed driving state; in addition, the number of consistent working characteristics of the preset agricultural machinery in the strip is more than that when turning at the headland.

[0098] Figure 8 shown is the variation of the driving speed of an agricultural machinery in a farmland over time, and the speed of the agricultural machinery is roughly distributed in the driving speed interval of 2.5 m / s and 4.5 m / s, therefore, the strip working speed interval of this farmland is the driving speed interval of 2.5 m / s to 4.5 m / s.

[0099] By determining the density distribution of speed track points, the track points in the peak interval are selected, and the built-in graphical display library function matplotlib of python is used to display the track points as Figure 8 shown, wherein the thick black points at both ends are non-interval track points, and the remaining non-thick ordinary five-pointed star points are track points selected by using the density function, that is, points whose speed is in the interval of 2.5 m / s to 4.5 m / s.

[0100] While agricultural machinery operates at a constant speed within the strips, the variable working conditions and complex road surfaces in the fields can cause drivers to accelerate or decelerate. This can result in the speed within the strip being similar to the speed at a turning point at the edge of the field. Consequently, points within the strip are sometimes misidentified as turning points, and vice versa. Outlier data was ultimately identified based on clustering results.

[0101] In step S235 of some embodiments, the outlier data is corrected according to the parameter correction strategy to obtain the corrected trajectory data. After executing step S234, which filters out outlier data whose values ​​of the operation trajectory point data fall within the correction interval based on the clustering results, the outlier data is corrected according to the preset parameter correction strategy in step S231 to obtain the corrected trajectory data. That is, outlier data obtained due to various practical reasons is corrected to obtain clean agricultural machinery turning trajectory data.

[0102] In some embodiments of the present invention, when the parameter correction strategy is a travel distance correction strategy, the agricultural machinery, during strip operations, travels at a constant speed, resulting in a uniform and abundant distance between adjacent trajectory points. However, when the agricultural machinery turns around, the distance between adjacent trajectory points changes; as the speed decreases, the distance becomes closer and smaller, and the distance characteristic becomes less obvious. Figure 9 The diagram shows how the distance a farm machine travels in a field changes over time. Figure 10 The corresponding figure is shown below. Figure 9 The driving distance density distribution map is shown. The initial screening method for the driving distance correction strategy is the same as that for the driving speed correction strategy, that is, by determining the density distribution of trajectory points, the corresponding peak interval is found, and the abnormal trajectory point data of driving distance in the strip are screened out.

[0103] The abnormal trajectory point data, i.e., U-turn trajectory points with similar driving distances, which were initially screened by the density function, were clustered using the DBSACAN clustering algorithm with adjusted parameters. The driving distance values ​​in the clustering results were then changed to the average inter-row distances. The corrected results are shown in Figure 11. Figure 11 The bolded black dots at both ends represent the turning trajectory data of the agricultural machinery, while the ordinary pentagrams and non-bold dots within the non-bolded strips represent the corrected trajectory data.

[0104] In some embodiments, reference Figure 12 As shown, step S240 may also include, but is not limited to, steps S241 to S243.

[0105] S241, compare the parameter value data corresponding to the operation trajectory point data with the correction interval range;

[0106] S242, if the parameter value data falls within the correction interval range, then the operation trajectory point data corresponding to the parameter value data is filtered out, and the filtered operation trajectory point data is used as the interval trajectory point data;

[0107] S243, Remove the corrected trajectory data from the interval trajectory point data to obtain the initial U-turn trajectory data.

[0108] In some embodiments, the parameter value data corresponding to the work trajectory point data is compared with the correction interval range. It is understood that by comparing the parameter value data of the work trajectory points with the correction interval range, if the parameter value data falls within the correction interval range, the work trajectory point data corresponding to the parameter value data is selected, and the selected work trajectory point data is used as the interval trajectory point data. Finally, the corrected trajectory data is removed from the interval trajectory point data to obtain the initial U-turn trajectory data.

[0109] In some embodiments, reference Figure 13 As shown, step S300 may also include, but is not limited to, steps S310 to S330.

[0110] S310, The initial U-turn trajectory data is processed according to the ten-fold cross-validation method to obtain the target parameters;

[0111] S320, Based on the random forest algorithm and the target parameters, the initial U-turn trajectory data is verified to obtain the verification result;

[0112] S330, if the verification value corresponding to the verification result exceeds the preset standard threshold, then the initial U-turn trajectory data is determined to be the target U-turn trajectory data.

[0113] In step S310 of some embodiments, the initial U-turn trajectory data is processed according to the ten-fold cross-validation method to obtain the target parameters.

[0114] Understandably, 10-fold cross-validation is used to test the accuracy of algorithms. It's a commonly used testing method. The dataset is divided into ten parts, and nine parts are used alternately as training data and one part as test data for trials. Each trial yields a corresponding accuracy (or error rate). The average of the accuracy (or error rate) of the ten trials is used as an estimate of the algorithm's precision. Generally, multiple 10-fold cross-validations (e.g., ten trials) are performed, and the average is calculated again as an estimate of the algorithm's accuracy.

[0115] In some embodiments, the optimal parameters of the algorithm, i.e. the target parameters, can be found using the data partitioning method of 10-fold cross-validation.

[0116] In step S320 of some embodiments, the initial U-turn trajectory data is verified according to the random forest algorithm and the target parameters to obtain the verification result.

[0117] The Random Forest algorithm, as we understand it, refers to a classifier that uses multiple trees to train and predict samples. N represents the number of training cases (samples), and M represents the number of features. The number of input features, m, is used to determine the decision outcome of a node in the decision tree; m should be much smaller than M. Then, samples are taken N times with replacement from the N training cases (samples) to form a training set (i.e., bootstrap sampling). The unsampled cases are used for prediction, and the error is evaluated. For each node, m features are randomly selected, and the decision of each node in the decision tree is based on these features. Based on these m features, the optimal splitting method is calculated.

[0118] Based on the random forest algorithm and the target parameters obtained in step S310, the initial U-turn trajectory data is verified to obtain the verification result, which is used to determine the target U-turn trajectory data.

[0119] In step S330 of some embodiments, if the verification value corresponding to the verification result exceeds a preset standard threshold, then the initial U-turn trajectory data is determined to be the target U-turn trajectory data. It can be understood that after executing step S320 to verify the initial U-turn trajectory data according to the random forest algorithm and the target parameters, and obtaining the verification result, if the verification value corresponding to the verification result exceeds a preset standard threshold, the initial U-turn trajectory data is determined to be the target U-turn trajectory data, thus demonstrating that the accuracy of identifying target U-turn trajectory data disclosed in this method is high.

[0120] In some embodiments, if the verification value corresponding to the verification result exceeds a preset standard threshold, then it is determined that the recognition accuracy of the initial U-turn trajectory data is relatively poor.

[0121] This invention provides a method for identifying the turning trajectory of agricultural machinery. The method involves acquiring historical operation trajectory data of a preset agricultural machinery performing a preset historical task; preprocessing the historical operation trajectory data to obtain preprocessed trajectory data; and then using a preset turning trajectory identification algorithm to identify the preprocessed trajectory data, thereby obtaining the target turning trajectory data corresponding to the preset agricultural machinery performing the preset historical task. This method can identify the turning trajectory data of agricultural machinery based on its historical operation trajectory data, allowing for further analysis and research of the turning trajectory data to improve the working efficiency of agricultural machinery during operations.

[0122] The following describes a device for identifying the turning trajectory of agricultural machinery provided by the present invention. The device for identifying the turning trajectory of agricultural machinery described below and the method for identifying the turning trajectory of agricultural machinery described above can be referred to in correspondence.

[0123] Reference Figure 14 As shown, the present invention also provides a device for recognizing the turning trajectory of agricultural machinery, comprising:

[0124] The acquisition module 1410 is used to acquire historical operation trajectory data when the preset agricultural machinery performs preset historical operations;

[0125] Preprocessing module 1420 is used to preprocess the historical operation trajectory data to obtain preprocessed trajectory data;

[0126] The identification module 1430 is used to identify and process the preprocessed trajectory data according to a preset U-turn identification algorithm to obtain the target U-turn trajectory data corresponding to the preset agricultural machinery when performing the preset historical operation.

[0127] According to the present invention, an agricultural machinery turning trajectory recognition device includes an acquisition module 1410, which is further configured to determine the agricultural machinery operation attribute information corresponding to the preset agricultural machinery based on preset requirements; and to acquire the historical operation trajectory data of the preset agricultural machinery when performing the preset historical operation from a preset agricultural machinery operation data storage system based on a preset system interface and the agricultural machinery operation attribute information.

[0128] According to the present invention, a preprocessing module 1420 for identifying agricultural machinery turning trajectories includes a preprocessing module further configured to: clean the historical operation trajectory data corresponding to each preset agricultural machinery to obtain standard operation trajectory data corresponding to each preset agricultural machinery; annotate the standard operation trajectory data to obtain annotated operation trajectory data corresponding to each preset agricultural machinery; correct the annotated operation trajectory data according to a preset parameter correction strategy to obtain corrected trajectory data; and determine the initial turning trajectory data based on the corrected trajectory data and the annotated operation trajectory data.

[0129] According to the present invention, a preprocessing module 1420 for identifying the turning trajectory of agricultural machinery is further configured to perform anomaly filtering processing on the historical operation trajectory data corresponding to each preset agricultural machinery based on a preset anomaly trajectory filtering strategy, thereby obtaining anomaly trajectory data corresponding to each preset agricultural machinery; and remove the anomaly trajectory data from the historical operation trajectory data to obtain the standard operation trajectory data corresponding to each preset agricultural machinery.

[0130] According to the present invention, a preprocessing module 1420 for identifying agricultural machinery turning trajectories includes at least: geographical location information; the marked operation trajectory data includes at least: operation trajectory point data; based on the geographical location information, it performs calculations on multiple operation trajectory point data respectively to obtain parameter value data corresponding to each operation trajectory point data; wherein the geographical location information is obtained by real-time sensing of the position information of the preset agricultural machinery when performing the preset historical operation by a preset sensor; wherein the operation trajectory point data is the data corresponding to the trajectory behavior generated by the preset agricultural machinery when performing the preset historical operation.

[0131] Based on the parameter correction strategy and the parameter value data, a correction interval range is determined for correcting multiple operation trajectory point data; wherein the correction interval range is the range range of parameter value data corresponding to the trajectory points generated when the preset agricultural machinery performs a turning behavior; the operation trajectory point data is clustered according to a preset density clustering algorithm and the setting parameters set on a preset development platform to obtain clustering results; based on the clustering results, outlier data whose values ​​of the operation trajectory point data are within the correction interval range are filtered out; the outlier data is corrected according to the parameter correction strategy to obtain the corrected trajectory data.

[0132] According to the present invention, a preprocessing module 1420 for identifying agricultural machinery turning trajectory is further configured to compare the parameter value data corresponding to the operation trajectory point data with the correction interval range; if the parameter value data falls within the correction interval range, the operation trajectory point data corresponding to the parameter value data is selected, and the selected operation trajectory point data is used as interval trajectory point data; the correction trajectory data is removed from the interval trajectory point data to obtain the initial turning trajectory data.

[0133] According to the present invention, an agricultural machinery turning trajectory recognition device includes a recognition module 1430, which is further configured to: the turning recognition algorithm includes at least: a random forest algorithm and a 10-fold cross-validation method; perform calculation processing on the initial turning trajectory data according to the 10-fold cross-validation method to obtain target parameters; perform verification processing on the initial turning trajectory data according to the random forest algorithm and the target parameters to obtain a verification result; if the verification value corresponding to the verification result exceeds a preset standard threshold, then determine the initial turning trajectory data as the target turning trajectory data.

[0134] This invention provides a device for recognizing agricultural machinery turning trajectories. It acquires historical operation trajectory data of a preset agricultural machinery performing a preset historical task; preprocesses the historical operation trajectory data to obtain preprocessed trajectory data; and then uses a preset turning recognition algorithm to identify the preprocessed trajectory data, obtaining the target turning trajectory data corresponding to the preset agricultural machinery performing the preset historical task. This device can identify the turning trajectory data of the agricultural machinery based on its historical operation trajectory data using the turning recognition algorithm, allowing for further analysis and research of the turning trajectory data to improve the working efficiency of the agricultural machinery during operations.

[0135] Figure 15 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 15As shown, the electronic device may include a processor 1510, a communications interface 1520, a memory 1530, and a communication bus 1540. The processor 1510, communications interface 1520, and memory 1530 communicate with each other via the communication bus 1540. The processor 1510 can call logical instructions in the memory 1530 to execute a method for identifying the turning trajectory of agricultural machinery. This method includes: acquiring historical operation trajectory data of a preset agricultural machinery performing a preset historical operation; preprocessing the historical operation trajectory data to obtain preprocessed trajectory data; and identifying the preprocessed trajectory data according to a preset turning recognition algorithm to obtain target turning trajectory data corresponding to the preset agricultural machinery performing the preset historical operation. Furthermore, the logical instructions in the memory 1530 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of the present invention, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of the present invention. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0136] On the other hand, the present invention also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer can execute a method for identifying the turning trajectory of agricultural machinery provided by the above methods. The method includes: acquiring historical operation trajectory data of a preset agricultural machinery performing a preset historical operation; preprocessing the historical operation trajectory data to obtain preprocessed trajectory data; and identifying the preprocessed trajectory data according to a preset turning recognition algorithm to obtain target turning trajectory data corresponding to the preset agricultural machinery performing the preset historical operation.

[0137] In another aspect, the present invention also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, implements a method for identifying the turning trajectory of agricultural machinery provided by the methods described above. The method includes: acquiring historical operation trajectory data of a preset agricultural machinery performing a preset historical operation; preprocessing the historical operation trajectory data to obtain preprocessed trajectory data; and identifying the preprocessed trajectory data according to a preset turning recognition algorithm to obtain target turning trajectory data corresponding to the preset agricultural machinery performing the preset historical operation.

[0138] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0139] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0140] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for recognizing the turning trajectory of agricultural machinery, characterized in that, include: Obtain historical operation trajectory data of agricultural machinery when it performs preset historical operations; The historical operation trajectory data is preprocessed to obtain preprocessed trajectory data; The preprocessed trajectory data includes at least: initial turning trajectory data; the preprocessing of the historical operation trajectory data to obtain preprocessed trajectory data includes: cleaning the historical operation trajectory data corresponding to each preset agricultural machine to obtain standard operation trajectory data corresponding to each preset agricultural machine; labeling the standard operation trajectory data to obtain labeled operation trajectory data corresponding to each preset agricultural machine; correcting the labeled operation trajectory data according to a preset parameter correction strategy to obtain corrected trajectory data; and determining the initial turning trajectory data based on the corrected trajectory data and the labeled operation trajectory data. The preprocessed trajectory data is processed by a preset U-turn recognition algorithm to obtain the target U-turn trajectory data corresponding to the preset agricultural machinery when performing the preset historical operation; wherein, the U-turn recognition algorithm includes at least: random forest algorithm and ten-fold cross-validation method; The step of identifying and processing the preprocessed trajectory data according to a preset U-turn recognition algorithm to obtain the target U-turn trajectory data corresponding to the preset agricultural machinery when performing the preset historical operation includes: The initial U-turn trajectory data is processed using the ten-fold cross-validation method to obtain target parameters; the initial U-turn trajectory data is then validated using the random forest algorithm and the target parameters to obtain validation results; if the validation value corresponding to the validation result exceeds a preset standard threshold, the initial U-turn trajectory data is determined to be the target U-turn trajectory data.

2. The method for identifying the turning trajectory of agricultural machinery according to claim 1, characterized in that, The acquisition of historical operation trajectory data when the preset agricultural machinery performs preset historical operations includes: The agricultural machinery operation attribute information corresponding to the preset agricultural machinery is determined according to the preset demand conditions; Based on the preset system interface and the agricultural machinery operation attribute information, the historical operation trajectory data of the preset agricultural machinery when performing the preset historical operation is obtained from the preset agricultural machinery operation data storage system.

3. The method for identifying the turning trajectory of agricultural machinery according to claim 1, characterized in that, The step of cleaning the historical operation trajectory data corresponding to each preset agricultural machine to obtain standard operation trajectory data corresponding to each preset agricultural machine includes: Based on the preset abnormal trajectory filtering strategy, the historical operation trajectory data corresponding to each preset agricultural machine is subjected to abnormal filtering processing to obtain the abnormal trajectory data corresponding to each preset agricultural machine. The abnormal trajectory data is removed from the historical operation trajectory data to obtain the standard operation trajectory data corresponding to each preset agricultural machine.

4. The method for identifying the turning trajectory of agricultural machinery according to claim 2, characterized in that, The agricultural machinery operation attribute information includes at least: geographic location information; The labeled work trajectory data includes at least: work trajectory point data; The step of correcting the labeled trajectory data according to a preset parameter correction strategy to obtain corrected trajectory data includes: Based on the geographic location information, the multiple operation trajectory point data are calculated and processed respectively to obtain the parameter value data corresponding to each operation trajectory point data; wherein the geographic location information is obtained by real-time sensing of the location information of the preset agricultural machinery when performing the preset historical operation by a preset sensor; wherein the operation trajectory point data is the data corresponding to the trajectory behavior generated by the preset agricultural machinery when performing the preset historical operation. Based on the parameter correction strategy and the parameter value data, a correction interval range is determined for correcting multiple operation trajectory point data; wherein the correction interval range is the interval range of parameter value data corresponding to the trajectory points generated when the preset agricultural machinery performs a turning behavior; The operation trajectory point data is clustered according to the preset density clustering algorithm and the set parameters in the preset development platform to obtain the clustering results; Based on the clustering results, outlier data points whose values ​​are within the correction interval are selected. The abnormal point data is corrected according to the parameter correction strategy to obtain the corrected trajectory data.

5. The method for identifying the turning trajectory of agricultural machinery according to claim 4, characterized in that, The step of determining the initial U-turn trajectory data based on the corrected trajectory data and the labeled operation trajectory data includes: The parameter values ​​corresponding to the operation trajectory point data are compared with the correction interval range. If the parameter value data falls within the correction interval, the operation trajectory point data corresponding to the parameter value data is filtered out, and the filtered operation trajectory point data is used as the interval trajectory point data. The corrected trajectory data is removed from the interval trajectory point data to obtain the initial U-turn trajectory data.

6. A device for recognizing the turning trajectory of agricultural machinery, characterized in that, include: The acquisition module is used to acquire historical operation trajectory data of agricultural machinery when it performs preset historical operations; The preprocessing module is used to preprocess the historical operation trajectory data to obtain preprocessed trajectory data; The preprocessed trajectory data includes at least: initial turning trajectory data; specifically, it is used for: cleaning the historical operation trajectory data corresponding to each of the preset agricultural machines to obtain standard operation trajectory data corresponding to each of the preset agricultural machines; labeling the standard operation trajectory data to obtain labeled operation trajectory data corresponding to each of the preset agricultural machines; correcting the labeled operation trajectory data according to a preset parameter correction strategy to obtain corrected trajectory data; and determining the initial turning trajectory data based on the corrected trajectory data and the labeled operation trajectory data. The identification module is used to identify the preprocessed trajectory data according to a preset U-turn identification algorithm to obtain the target U-turn trajectory data corresponding to the preset agricultural machinery performing the preset historical operation; wherein, the U-turn identification algorithm includes at least: random forest algorithm and ten-fold cross-validation method; the identification module is specifically used to: perform calculation processing on the initial U-turn trajectory data according to the ten-fold cross-validation method to obtain target parameters; perform verification processing on the initial U-turn trajectory data according to the random forest algorithm and the target parameters to obtain verification results; if the verification value corresponding to the verification result exceeds a preset standard threshold, then the initial U-turn trajectory data is determined to be the target U-turn trajectory data.

7. An electronic device comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps of the agricultural machinery turning trajectory recognition method as described in any one of claims 1 to 5.

8. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the steps of the agricultural machinery turning trajectory recognition method as described in any one of claims 1 to 5.

Images