Satellite positioning track processing method and device, equipment and storage medium

Abstract

The invention discloses a satellite positioning track processing method and device, equipment and a storage medium. The method comprises the following steps: acquiring first track data of satellite positioning reported by at least one electronic device; preprocessing the first trajectory data to obtain preprocessed second trajectory data; performing classification prediction on the second trajectory data to obtain at least one classification result corresponding to the second trajectory data; under the condition that at least one abnormal track classification exists in the at least one classification result, track correction processing is carried out on each abnormal track classification, and a corrected target track is obtained; the processing flow can be effectively simplified, the abnormal trajectory types can be flexibly increased, and a new abnormal trajectory scene can be covered more easily.

Description

Technical Field

[0001] This application relates to the field of computer technology, and in particular, to a method, apparatus, device, and storage medium for processing satellite positioning trajectories. Background Art

[0002] In the related art, at locations such as near high-rise buildings, underground buildings (such as underground parking lots, subways, or tunnels), and inside buildings, due to reasons such as wall refraction or indoor signal attenuation, the signal error increases, which easily causes inaccurate positioning of the Global Positioning System (GPS). For this problem, there is currently no effective solution. Summary of the Invention

[0003] To solve the related technical problems, embodiments of this application provide a method, apparatus, device, and storage medium for processing satellite positioning trajectories, which can correct the trajectories with GPS positioning errors to make GPS positioning more accurate.

[0004] To achieve the above object, the technical solution of the embodiments of this application is implemented as follows:

[0005] Embodiments of this application provide a method for processing satellite positioning trajectories, the method including:

[0006] Obtain first trajectory data of satellite positioning reported by at least one electronic device;

[0007] Perform preprocessing on the first trajectory data to obtain second trajectory data after preprocessing;

[0008] Perform classification prediction on the second trajectory data to obtain at least one classification result corresponding to the second trajectory data;

[0009] In the case where there is at least one abnormal trajectory classification among the at least one classification result, perform trajectory correction processing on each abnormal trajectory classification to obtain a corrected target trajectory.

[0010] In the above solution, the performing preprocessing on the first trajectory data to obtain second trajectory data after preprocessing includes:

[0011] Perform screening processing on the first trajectory data to obtain target data corresponding to the first trajectory data;

[0012] Determine the second trajectory data based on the attribute parameters of the target data.

[0013] In the above solution, the attribute parameters at least include longitude parameters, latitude parameters, speed parameters, and azimuth parameters; the determining the second trajectory data based on the attribute parameters of the target data includes:

[0014] Determine the second trajectory data according to the longitude parameter, the latitude parameter, the speed parameter and the azimuth parameter in chronological order.

[0015] In the above solution, the classifying and predicting the second trajectory data to obtain at least one classification result corresponding to the second trajectory data includes:

[0016] Input the second trajectory data into a preset classification model to obtain the at least one classification result corresponding to the second trajectory data.

[0017] In the above solution, the method further includes:

[0018] Obtain the initial trajectory data of satellite positioning reported by at least one of the electronic devices;

[0019] Input the initial trajectory data into an initial classification model for training to obtain the preset classification model.

[0020] In the above solution, the abnormal trajectory classification includes a first abnormal trajectory classification; the first abnormal trajectory classification indicates that there are at least two abnormal points around the end point of the satellite positioning trajectory; in the case that there is at least one abnormal trajectory classification in the at least one classification result, performing trajectory correction processing on each abnormal trajectory classification to obtain a corrected target trajectory, including:

[0021] In the case that the first abnormal trajectory classification exists in the at least one classification result, obtain a first parameter corresponding to a first abnormal point and a second parameter corresponding to a second abnormal point among the at least two abnormal points;

[0022] Obtain the corrected target trajectory according to the first parameter and the second parameter.

[0023] In the above solution, the abnormal trajectory includes a second abnormal trajectory classification; the second abnormal trajectory classification indicates that there is at least one folded-back trajectory in the satellite positioning trajectory; in the case that there is at least one abnormal trajectory classification in the at least one classification result, performing trajectory correction processing on each abnormal trajectory classification to obtain a corrected target trajectory, including:

[0024] In the case that the second abnormal trajectory classification exists in the at least one classification result, obtain a third parameter corresponding to a first point in the satellite positioning trajectory; the first point represents a normal point in the satellite positioning trajectory;

[0025] Obtain a fourth parameter corresponding to a second point and the first point according to the third parameter; the second point represents other points in the satellite positioning trajectory except the first point;

[0026] Obtain the corrected target trajectory based on the fourth parameter.

[0027] In the above solution, the abnormal trajectory includes a third abnormal trajectory classification; the third abnormal trajectory classification represents that the distance between the third point and the fourth point in the satellite positioning trajectory data is greater than a preset threshold; the third point and the fourth point are any points in the satellite positioning trajectory; in the case that there is at least one abnormal trajectory classification in the at least one classification result, performing a trajectory correction process on each abnormal trajectory classification to obtain a corrected target trajectory, including:

[0028] In the case that the third abnormal trajectory classification exists in the at least one classification result, obtain a fifth point in the satellite positioning trajectory; the fifth point is a normal point in the satellite positioning trajectory data;

[0029] Replace the third point or the fourth point with the fifth point to obtain the corrected target trajectory.

[0030] An embodiment of the present application further provides a satellite positioning trajectory processing device, and the device includes:

[0031] An acquisition unit, configured to acquire first trajectory data of satellite positioning reported by at least one electronic device;

[0032] A preprocessing unit, configured to preprocess the first trajectory data to obtain preprocessed second trajectory data;

[0033] A classification and prediction unit, configured to perform classification and prediction on the second trajectory data to obtain at least one classification result corresponding to the second trajectory data;

[0034] A correction processing unit, configured to perform a trajectory correction process on each abnormal trajectory classification in the case that there is at least one abnormal trajectory classification in the at least one classification result, to obtain a corrected target trajectory.

[0035] An embodiment of the present application further provides a satellite positioning trajectory processing device, including: a processor and a memory for storing a computer program that can run on the processor,

[0036] wherein, when the processor is used to run the computer program, it executes any step of the above method.

[0037] An embodiment of the present application further provides a storage medium, on which a computer program is stored, and when the computer program is executed by a processor, any step of the above method is implemented.

[0038] The satellite positioning trajectory processing method, device, equipment, and storage medium provided by the embodiments of this application. Among them, the method includes: obtaining first trajectory data of satellite positioning reported by at least one electronic device; preprocessing the first trajectory data to obtain second trajectory data after preprocessing; classifying and predicting the second trajectory data to obtain at least one classification result corresponding to the second trajectory data; in the case that there are at least one abnormal trajectory classifications among the at least one classification results, performing trajectory correction processing on each abnormal trajectory classification to obtain corrected target trajectories. The technical solution of the embodiments of this application realizes precise GPS positioning by classifying and predicting the trajectory data of satellite positioning reported by electronic devices and performing correction processing on abnormal trajectories when abnormal trajectory classifications occur (i.e., GPS positioning has errors). BRIEF DESCRIPTION OF THE DRAWINGS

[0039] Figure 1 It is a schematic diagram of the flow of a satellite positioning trajectory processing method provided by the embodiments of this application;

[0040] Figure 2 It is a schematic diagram of a type of abnormal trajectory provided by the embodiments of this application;

[0041] Figure 3 It is a schematic diagram of the GPS abnormal trajectory classification processing flow provided by the embodiments of this application;

[0042] Figure 4 It is a schematic diagram of a deep learning model provided by the embodiments of this application;

[0043] Figure 5 It is a schematic diagram of a satellite positioning trajectory processing device provided by the embodiments of this application;

[0044] Figure 6 It is a schematic diagram of a hardware entity structure of a satellite positioning trajectory processing device provided by the embodiments of this application. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0045] The present invention will be further described in detail below with reference to the accompanying drawings and specific embodiments.

[0046] According to the GPS positioning principle, GPS measurement errors are diverse and can be divided into errors related to GPS satellites, errors related to signal propagation, errors related to receiving devices, etc. In daily life, the situation of inaccurate GPS positioning often occurs near some high-rise buildings, underground buildings (such as underground parking lots, subways or tunnels), and inside buildings. Due to reasons such as wall refraction and indoor signal attenuation, the signal error increases, resulting in inaccurate positioning. In addition, the GPS mobile receiver needs to transmit data to the background server through a wireless network. In some areas with weak wireless signal coverage, the network transmission is unstable, resulting in data output delay or even data loss, which will also cause a large error in the GPS position information received by the background and displayed on the map.

[0047] GPS errors will bring many adverse effects to the application of GPS data. For example, in the management of a company's fleet, if the original GPS data is directly displayed on the map, it will be found that there are many obvious mismatches with the actual driving trajectory; the driving mileage calculated based on GPS data often exceeds the true value by a large margin; false warning messages such as "crossing the border" and "speeding" often appear. In order to improve GPS errors, a large number of studies have also been carried out.

[0048] (1) Cache GPS sampling points starting from the effective starting sampling point of GPS, and delete error points during the caching process. After screening the cached GPS sampling points, send them to the Geographic Information System (GIS) server. The GIS server generates a trajectory vector map based on the screened GPS sampling points. Since operations such as error point deletion and screening are performed on the GPS sampling points, the GIS server generates a trajectory vector map, realizing the optimization of the GPS trajectory.

[0049] (2) Judge whether the currently parsed GPS data is valid by comparing three parameters: First, compare the parsed GPS data with the base station positioning data; second, compare the speed data after GPS parsing with the actual data inside the vehicle; third, compare the vehicle direction changes at consecutive moments. If it is valid, retain the data; if it is invalid, filter the data.

[0050] (3) By obtaining the target trajectory to be processed; respectively performing filtering and deviation correction on each position point in the target trajectory to obtain the first corrected trajectory; calculating the actual distance between two points at adjacent times in the first corrected trajectory, and calculating the limit distance between two points at adjacent times in the first corrected trajectory; eliminating abnormal position points from the first corrected trajectory according to the actual distance and the limit distance to obtain the second corrected trajectory; respectively performing filtering and deviation correction on each position point in the second corrected trajectory to obtain the third corrected trajectory. The whole process can solve the problem of large trajectory errors during vehicle operation.

[0051] (4) Preprocess the collected GPS trajectory data to obtain the second GPS trajectory data, and then overlay the second GPS trajectory data with the electronic map trajectory to generate the overlay trajectory data. The overlay trajectory data includes a trajectory adaptive ratio, and the map level parameter is determined according to the adaptive ratio fed back by the overlay trajectory data. The map level parameter is used to reflect the travel distance in multiple travel modes. Thus, the purpose of trajectory classification is achieved, and the travel modes (such as airplane, train, car, walking, etc.) of the end user can be effectively classified.

[0052] In current GPS trajectory optimization technologies, basically, the GPS trajectory is optimized by logical judgment to filter and delete invalid points or error points. Some optimization methods have relatively simple logical judgment conditions and cover some GPS abnormal trajectory scenarios, and can only achieve partial correction effects. There are also some optimization methods. In order to cover more abnormal trajectory scenarios, the logical judgment is complex and the calculation complexity is high. In addition, when error points continuously appear during vehicle movement, if all the error points are deleted, the two effective points before and after will be pulled into a straight line on the map, resulting in a situation where the optimized driving trajectory does not match the road in the map.

[0053] For example, the validity of the currently parsed GPS data is judged by comparing three parameters: First, compare the parsed GPS data with the base station positioning data; Second, compare the speed data after GPS parsing with the actual data inside the vehicle; Third, compare the vehicle direction changes at consecutive times. If it is valid, the data is retained; if it is invalid, the data is filtered. Because the base station positioning accuracy is not high and there are some areas with weak wireless signals, the first judgment condition has some limitations and is only suitable for making a rough-grained judgment. The second and third conditions require the vehicle internal driving speed data and the steering wheel turning data, and a vehicle front-mounted Internet of Vehicles device is required.

[0054] For another example, by judging the effective starting point, GPS trajectory points are sampled at a certain period, and whether the sampled value is effective is judged according to the comparison between the distance between adjacent sampled points and the first threshold value. In addition, the judgment of inflection points is also made, and then a trajectory vector map is generated and matched with the GIS map. This method can optimize the GPS trajectory to a certain extent and the computational complexity is not high. However, in the face of some disordered static or dynamic drift points in the middle of the GPS trajectory, the judgment logic is too simple and cannot cover the relevant processing.

[0055] Based on this, an embodiment of the present application provides a method for processing satellite positioning trajectories. First, abnormal trajectories are classified, and then corresponding processing is performed according to the classification type, so as to achieve the goal of optimizing abnormal trajectories.

[0056] Figure 1 FIG. is a schematic diagram of a method for processing satellite positioning trajectories provided by an embodiment of the present application; as Figure 1 shown, the method includes:

[0057] S101: Obtain first trajectory data of satellite positioning reported by at least one electronic device;

[0058] S102: Preprocess the first trajectory data to obtain second trajectory data after preprocessing;

[0059] S103: Classify and predict the second trajectory data to obtain at least one classification result corresponding to the second trajectory data;

[0060] S104: In the case that there is at least one abnormal trajectory classification in the at least one classification result, perform trajectory correction processing on each abnormal trajectory classification to obtain a corrected target trajectory.

[0061] It should be noted that the satellite positioning trajectory processing method can be exemplified as a GPS trajectory processing method in practical applications.

[0062] In S101, the electronic device can be understood as any device that can collect trajectory data of satellite positioning, which is not limited herein. As an example, the electronic device can be a terminal, such as an in-vehicle computer or a mobile phone. The first trajectory data can be understood as the original GPS trajectory data.

[0063] The obtaining of the first trajectory data of satellite positioning reported by at least one electronic device can be exemplified as obtaining the original GPS trajectory data reported by a terminal.

[0064] In S102, the second trajectory data can be understood as the trajectory data obtained by preprocessing the original GPS trajectory data.

[0065] The preprocessing of the first trajectory data to obtain the preprocessed second trajectory data can be understood as screening the first trajectory data to obtain the target data corresponding to the first trajectory data; and determining the second trajectory data based on the attribute parameters of the target data.

[0066] In S103, the classification prediction of the second trajectory data to obtain at least one classification result corresponding to the second trajectory data can be understood as inputting the second trajectory data into a preset classification model to obtain the at least one classification result corresponding to the second trajectory data.

[0067] It should be noted that the at least one classification result includes a normal trajectory and at least one abnormal trajectory classification.

[0068] In S104, the at least one abnormal trajectory classification at least includes a first abnormal trajectory classification, a second abnormal trajectory classification, and a third abnormal trajectory classification. The first abnormal trajectory classification indicates that there are at least two abnormal points around the end point of the satellite positioning trajectory; the second abnormal trajectory classification indicates that there is at least one trajectory of turning back in the satellite positioning trajectory; the third abnormal trajectory classification indicates that the distance between the third point and the fourth point in the satellite positioning trajectory data is greater than a preset threshold, and the third point and the fourth point are any points in the satellite positioning trajectory.

[0069] Among them, the classification result is represented by the output Y value. 1 represents the first abnormal trajectory classification, and in practical applications, it can be exemplified as a static drifting abnormal trajectory; 2 represents the second abnormal trajectory classification, and in practical applications, it can be exemplified as a repeatedly turning back abnormal trajectory; 3 represents the third abnormal trajectory classification, and in practical applications, it can be exemplified as a dynamic drifting abnormal trajectory. Figure 2 The following is a schematic diagram of an abnormal trajectory type provided by an embodiment of the present application, as Figure 2 shown Figure 2 in (a) represents a static drifting abnormal trajectory; Figure 2 in (b) represents a repeatedly turning back abnormal trajectory; Figure 2 in (c) represents a dynamic drifting abnormal trajectory.

[0070] It should be noted that in the case where there is a normal trajectory classification in the at least one classification result, the second trajectory data is used as the target trajectory data. As an example, in the case where the trajectory is classified as a normal trajectory category, no processing is performed, and the input trajectory points are output as they are.

[0071] The present application first classifies the trajectory data. After obtaining the classification result, the abnormal trajectory classification among them is corrected, which can simplify the processing flow of the GPS trajectory correction algorithm and improve the accuracy of the correction result.

[0072] In one embodiment, the preprocessing of the first trajectory data to obtain the preprocessed second trajectory data includes:

[0073] Performing screening processing on the first trajectory data to obtain target data corresponding to the first trajectory data;

[0074] Determining the second trajectory data based on the attribute parameters of the target data.

[0075] Among them, the performing screening processing on the first trajectory data to obtain target data corresponding to the first trajectory data can be understood as screening out target data from the original GPS data. The attribute parameters of the target data at least include longitude parameter, latitude parameter, speed parameter, and azimuth parameter.

[0076] The determining the second trajectory data based on the attribute parameters of the target data can be understood as determining the second trajectory data in chronological order of the longitude parameter, the latitude parameter, the speed parameter, and the azimuth parameter.

[0077] In one embodiment, the attribute parameters at least include longitude parameter, latitude parameter, speed parameter, and azimuth parameter; the determining the second trajectory data based on the attribute parameters of the target data includes:

[0078] Determining the second trajectory data in chronological order of the longitude parameter, the latitude parameter, the speed parameter, and the azimuth parameter.

[0079] Among them, the determining the second trajectory data in chronological order of the longitude parameter, the latitude parameter, the speed parameter, and the azimuth parameter can be exemplified as storing the GPS data in sequence according to the time series of four dimensions of longitude, latitude, speed, and azimuth to determine the second trajectory data.

[0080] In one embodiment, the classifying and predicting the second trajectory data to obtain at least one classification result corresponding to the second trajectory data includes:

[0081] Inputting the second trajectory data into a preset classification model to obtain the at least one classification result corresponding to the second trajectory data.

[0082] Among them, the preset classification model is a trained classification model, and the type of the model can be determined according to the actual situation and is not limited here. As an example, the preset classification model can be a trained deep learning classification prediction model of Convolutional Neural Network (CNN).

[0083] Inputting the second trajectory data into a preset classification model to obtain the at least one classification result corresponding to the second trajectory data; this can be illustrated by an example: inputting the second trajectory data into a preset classification model, and outputting Y as the trajectory type value, where 0 represents a normal trajectory, 1 represents a static drifting anomaly trajectory, 2 represents a repeated turning-back anomaly trajectory, and 3 represents a dynamic drifting anomaly trajectory, a total of four types.

[0084] In one embodiment, the method further includes:

[0085] Obtaining initial trajectory data of satellite positioning reported by at least one of the electronic devices;

[0086] Inputting the initial trajectory data into an initial classification model for training to obtain the preset classification model.

[0087] Among them, the obtaining of the initial trajectory data of satellite positioning reported by at least one of the electronic devices; inputting the initial trajectory data into an initial classification model for training to obtain the preset classification model; this can be illustrated by an example: obtaining the initial trajectory data reported by an electronic device, classifying the initial trajectory data into four types: the above-mentioned normal trajectory, static drifting anomaly trajectory, repeated turning-back anomaly trajectory, and dynamic drifting anomaly trajectory, and inputting the classified initial trajectory data into a deep learning model for training to obtain a trained deep learning classification prediction model.

[0088] In the embodiment of the present application, the abnormal trajectory scenarios can be flexibly added. After adding a new abnormal trajectory scenario, it is only necessary to retrain the data of the newly added abnormal trajectory scenario, and it is easier to cover the new abnormal trajectory scenario.

[0089] In one embodiment, the abnormal trajectory classification includes a first abnormal trajectory classification; the first abnormal trajectory classification indicates that there are at least two abnormal points around the end point of the satellite positioning trajectory; in the case where there is at least one abnormal trajectory classification among the at least one classification result, performing trajectory correction processing on each abnormal trajectory classification to obtain a corrected target trajectory, including:

[0090] In the case where the first abnormal trajectory classification exists among the at least one classification result, obtaining a first parameter corresponding to a first abnormal point and a second parameter corresponding to a second abnormal point among the at least two abnormal points;

[0091] Obtaining the corrected target trajectory according to the first parameter and the second parameter.

[0092] Among them, the first abnormal trajectory can be understood as a static drifting anomaly trajectory. The abnormal point can be understood as a drifting point, and the drifting point refers to a point with inaccurate position due to various reasons (such as signal interference or building occlusion, etc.).

[0093] The first parameter corresponding to the first anomaly point can be understood as the time series value of the first drift point; the second parameter corresponding to the second anomaly point can be understood as the time series value of the second drift point.

[0094] The corrected target trajectory obtained according to the first parameter and the second parameter can be understood as calculating the center point between the first anomaly point and the second anomaly point according to the first parameter and the second parameter; replacing the first anomaly point and the second anomaly point with the center point to obtain the target trajectory.

[0095] In one embodiment, the abnormal trajectory includes a second abnormal trajectory classification; the second abnormal trajectory classification represents a trajectory with at least one U-turn in the satellite positioning trajectory; in the case where at least one abnormal trajectory classification exists in the at least one classification result, performing trajectory correction processing on each abnormal trajectory classification to obtain a corrected target trajectory, including:

[0096] In the case where the second abnormal trajectory classification exists in the at least one classification result, obtaining a third parameter corresponding to a first point in the satellite positioning trajectory; the first point represents a normal point in the satellite positioning trajectory;

[0097] Obtaining a fourth parameter corresponding to a second point and the first point according to the third parameter; the second point represents other points in the satellite positioning trajectory except the first point;

[0098] Obtaining the corrected target trajectory based on the fourth parameter.

[0099] Among them, the second abnormal trajectory classification can be understood as a repeatedly U-turn abnormal trajectory classification. The normal point can be understood as a valid point.

[0100] The third parameter corresponding to the first point can be understood as the position parameter of the first point. The fourth parameter can be understood as the distance parameter between the second point and the first point.

[0101] When the second abnormal trajectory classification exists in the at least one classification result, obtain a third parameter corresponding to a first point in the satellite positioning trajectory; the first point represents a normal point in the satellite positioning trajectory; obtain a fourth parameter corresponding to between a second point and the first point according to the third parameter; the second point represents other points in the satellite positioning trajectory except the first point; obtain the corrected target trajectory based on the fourth parameter; it can be illustrated by an example that when the trajectory type is a repeatedly turning-back abnormal trajectory category, calculate the distance between each point in the original trajectory and the previous valid point to obtain a distance set D, sort the distance set D in ascending order to obtain D′, where D′ represents a new distance set obtained by re-sorting the distances in the distance set D in ascending order, and output the corrected trajectory, that is, the target trajectory, according to the new subscript of D′.

[0102] In one embodiment, the abnormal trajectory includes a third abnormal trajectory classification; the third abnormal trajectory classification represents that the distance between a third point and a fourth point in the satellite positioning trajectory data is greater than a preset threshold; the third point and the fourth point are any points in the satellite positioning trajectory; when at least one abnormal trajectory classification exists in the at least one classification result, perform trajectory correction processing on each abnormal trajectory classification to obtain a corrected target trajectory, including:

[0103] When the third abnormal trajectory classification exists in the at least one classification result, obtain a fifth point in the satellite positioning trajectory; the fifth point is a normal point in the satellite positioning trajectory data;

[0104] Replace the third point or the fourth point with the fifth point to obtain the corrected target trajectory.

[0105] The third abnormal trajectory classification can be understood as a dynamic drifting abnormal trajectory category. The third abnormal trajectory classification represents that the distance between a third point and a fourth point in the satellite positioning trajectory data is greater than a preset threshold, and it can be illustrated by an example that an abnormal GPS point drifts to a place several kilometers away from the actual position; where the preset threshold can be determined according to the actual situation.

[0106] When the third abnormal trajectory classification exists in the at least one classification result, obtain a fifth point in the satellite positioning trajectory; the fifth point is a normal point in the satellite positioning trajectory data; replace the third point or the fourth point with the fifth point to obtain the corrected target trajectory; it can be illustrated by an example that replace the drifting point with the previous valid point to obtain the target trajectory.

[0107] For ease of understanding, here is an example of a GPS trajectory optimization method based on abnormal trajectory classification. By using a deep learning convolutional neural network, the GPS trajectory is first predicted and classified into four categories: the first category is the normal trajectory; the second category is the static drift abnormal trajectory type; the third category is the repeated turning-back abnormal trajectory type; the fourth category is the dynamic drift abnormal trajectory type. Then, according to the classification results, targeted logical calculations are performed.

[0108] Among them, GPS abnormal trajectories can generally be summarized into three categories: the first category is the static drift abnormal trajectory type when the vehicle is stationary or moving at a low speed. After the GPS mobile receiver reaches the end point, a large number of drift points near the end point are received; the second category is the repeated turning-back abnormal trajectory type during the vehicle's driving process, where the GPS trajectory shows that the GPS terminal makes repeated turning-back movements; the third category is the dynamic drift abnormal trajectory type during the vehicle's driving process, where suddenly an abnormal GPS point drifts to a place several kilometers away from the actual position.

[0109] Figure 3 This is a schematic diagram of the GPS abnormal trajectory classification processing flow provided by the embodiments of the present application. As Figure 3 shown, the preprocessed GPS data is input into the CNN deep learning network to obtain classification prediction results, such as normal trajectories, static drift abnormal trajectories, repeated turning-back abnormal trajectories, or dynamic drift abnormal trajectories; then, personalized algorithm processing is performed according to the classification results, and finally the corrected GPS data is output. Among them, the GPS data is processed by the data preprocessing unit, the classification prediction is performed by the deep learning network unit, and the personalized algorithm processing is performed by the personalized algorithm unit.

[0110] The data preprocessing unit is used to screen, organize, and store the data; the CNN deep learning network unit: is used for the trained model to perform classification prediction; the personalized algorithm unit is used to perform targeted logical calculations on the predicted types.

[0111] Data preprocessing unit: In order to fully reflect various characteristics of the GPS trajectory, four representative parameters of longitude, latitude, speed, and azimuth are selected from the original GPS data in this application. And the GPS data is stored in sequence according to the time series of the four dimensions of longitude, latitude, speed, and azimuth. A large amount of pre-classified data is required for the training process of the classification prediction model, and the input data needs to be of the same length, which can be defined as T. The principle adopted is "cutting off the long and making up the short". If the length is greater than T, it is truncated from the end of the data; if the data is less than T, the number 0 is supplemented at the end of the data.

[0112] Deep learning network unit: The CNN deep learning network is used to collect four types of trajectories, namely daily normal trajectories, static drifting abnormal trajectories, repeated turning-back abnormal trajectories, and dynamic drifting abnormal trajectories. After training, a CNN deep learning classification prediction model is generated. Input definition: The input is a time series composed of four-dimensional variables of longitude, latitude, speed, and azimuth. X = (X 1 , X 2 ..., X t ) represents the time series values of length t, where the values of the four dimensions of longitude, latitude, speed, and azimuth at the i-th moment are represented. Figure 4 This is a schematic diagram of a deep learning model provided by an embodiment of the present application. As Figure 4 shown, Figure 4 it contains five layers. The output (output) is obtained from the input (input) through the first layer, the second layer, the third layer, the fourth layer, and the fifth layer. The first layer and the second layer are a convolutional layer + activation function + pooling layer (Conv + ReLU + MaxPool2d); the third layer is a fully connected layer including an activation function (Fc1(nn.Linear + ReLU)); the fourth layer is a fully connected layer including an activation function (Fc2(nn.Linear + ReLU)); the fifth layer is a fully connected layer (Fc3(nn.Linear)).

[0113] (1) Convolutional layer: The two-dimensional convolution (Conv2d) is used for convolution operations. For example, Output = nn.Conv2d(in_channels, out_channels, kernel_size), where in_channels is the number of input parameter channels, out_channels is the number of channels generated by convolution, and kernel_size is the convolution kernel size, which is a tuple of type (int, int) in this application. For example, (2, 3) is a convolution kernel with a height of 2 and a width of 3.

[0114] (2) Activation function: The rectified linear unit (ReLU) is used. output_P = nn.ReLU(inplace = False)(input). ReLU increases the non-linear relationship between the layers of the neural network, better solves the problem of gradient disappearance, and does not perform exponential operations, reducing the complexity of the algorithm.

[0115] (3) Pooling layer: sandwiched between consecutive convolutional layers, used to compress the amount of data and parameters. It can be expressed as: output_P = nn.MaxPool2d(kernel_size, stride), where kernel_size can be regarded as a sliding window, and the window size is defined by the user. Stride determines how this window slides, and the default stride is the same as the maximum pooling window size.

[0116] (4) Fully Connected Layer: a linear layer in a neural network, which can be expressed as follows: Output_F = nn.Linear(in_features, out_features), where in_features represents the size of the input sample, and out_features represents the size of the output sample.

[0117] Output definition: The output Y is a numerical value of the trajectory type. 0 represents a normal trajectory, 1 represents a static drifting anomaly trajectory, 2 represents a repeated turning-back anomaly trajectory, and 3 represents a dynamic drifting anomaly trajectory, a total of four types.

[0118] Personalized algorithm unit: Facing different classification results, make targeted logical algorithm corrections. Each classification result has its own independent set of logical algorithms for processing. According to the output Y value, the processing is as follows:

[0119] (1) If Y is equal to 0, the trajectory type is judged as the normal trajectory category. The processing principle is not to process, and the input trajectory points are output as they are.

[0120] (2) If the output Y is equal to 1, the trajectory type is judged as the static drifting anomaly trajectory type. The processing principle is to calculate the center point X of these static drifting points center , and replace the remaining static drifting points with the center point; among them, X center = FunCenterPoint(X 1 , X 2 , …, X t ).

[0121] (3) If Y is equal to 2, the trajectory type is judged as the repeated turning-back anomaly trajectory type. The processing principle is that after reordering according to the distance from the previous valid point, a smooth GPS trajectory can be obtained. First, calculate the distance between each point in the original trajectory and the previous valid point to obtain a distance set D, and then sort the distance set D in ascending order to obtain D' = Sort(D), and output the corrected trajectory X' according to the new subscript of D'.

[0122] (4) If Y is equal to 3, the trajectory type is judged as the dynamic drifting anomaly trajectory category. The processing principle is to replace the drifting points with the previous valid point.

[0123] In the embodiments of the present application, the type of the trajectory is first determined, and then corresponding logical processing is performed according to different trajectories, which can cover more abnormal trajectory scenarios, achieve better optimization effects, and have lower algorithm complexity.

[0124] To implement the method of the embodiments of the present application, the embodiments of the present application also provide a satellite positioning trajectory processing device. Figure 5 As shown in the schematic diagram of a satellite positioning trajectory processing device provided by the embodiments of the present application, Figure 5 as shown, the device 500 includes:

[0125] An acquisition unit 501, configured to acquire first trajectory data of satellite positioning reported by at least one electronic device;

[0126] A preprocessing unit 502, configured to preprocess the first trajectory data to obtain second trajectory data after preprocessing;

[0127] A classification and prediction unit 503, configured to perform classification and prediction on the second trajectory data to obtain at least one classification result corresponding to the second trajectory data;

[0128] A correction processing unit 504, configured to perform trajectory correction processing on each abnormal trajectory classification in the case that there is at least one abnormal trajectory classification in the at least one classification result, to obtain a corrected target trajectory.

[0129] In one embodiment, the preprocessing unit 502 is further configured to perform screening processing on the first trajectory data to obtain target data corresponding to the first trajectory data; and determine the second trajectory data based on the attribute parameters of the target data.

[0130] In one embodiment, the preprocessing unit 502 is further configured to determine the second trajectory data in chronological order of the longitude parameter, the latitude parameter, the speed parameter, and the azimuth parameter.

[0131] In one embodiment, the classification and prediction unit 503 is further configured to input the second trajectory data into a preset classification model to obtain the at least one classification result corresponding to the second trajectory data.

[0132] In one embodiment, the device 500 further includes a training unit, configured to acquire initial trajectory data of satellite positioning reported by at least one of the electronic devices; and input the initial trajectory data into an initial classification model for training to obtain the preset classification model.

[0133] In one embodiment, the correction processing unit 504 is further configured to, when the first abnormal trajectory classification exists in the at least one classification result, obtain a first parameter corresponding to a first abnormal point and a second parameter corresponding to a second abnormal point among the at least two abnormal points; and obtain the corrected target trajectory according to the first parameter and the second parameter.

[0134] In one embodiment, the correction processing unit 504 is further configured to, when the second abnormal trajectory classification exists in the at least one classification result, obtain a third parameter corresponding to a first point in the satellite positioning trajectory; the first point represents a normal point in the satellite positioning trajectory; obtain a fourth parameter corresponding to between a second point and the first point according to the third parameter; the second point represents other points in the satellite positioning trajectory except the first point; and obtain the corrected target trajectory based on the fourth parameter.

[0135] In one embodiment, the correction processing unit 504 is further configured to, when the third abnormal trajectory classification exists in the at least one classification result, obtain a fifth point in the satellite positioning trajectory; the fifth point is a normal point in the satellite positioning trajectory data; and replace the third point or the fourth point with the fifth point to obtain the corrected target trajectory.

[0136] It should be noted that the satellite positioning trajectory processing device and the satellite positioning trajectory processing method provided in the above embodiments belong to the same concept. For the specific implementation process, please refer to the method embodiments, which will not be elaborated here.

[0137] Based on the hardware implementation of the above program modules, an embodiment of the present application further provides a satellite positioning trajectory processing device, including a memory and a processor. The memory stores a computer program that can run on the processor, and when the processor executes the program, it implements the steps in the satellite positioning trajectory processing method provided in the above embodiments.

[0138] Correspondingly, an embodiment of the present application provides a computer-readable storage medium, on which a computer program is stored. When the computer program is executed by a processor, it implements the steps in the satellite positioning trajectory processing method provided in the above method embodiments.

[0139] It should be pointed out here that the descriptions of the above storage medium and device embodiments are similar to the descriptions of the above method embodiments, and have beneficial effects similar to those of the method embodiments. For the technical details not disclosed in the storage medium and device embodiments of the present application, please refer to the descriptions of the method embodiments of the present application for understanding.

[0140] It should be noted that Figure 6 is a schematic diagram of a hardware entity structure of the satellite positioning trajectory processing device provided in the embodiment of the present application, asFigure 6 As shown, the hardware entities of the satellite positioning trajectory processing device 600 include: a processor 601 and a memory 602. Optionally, the satellite positioning trajectory processing device 600 may further include a communication interface 603.

[0141] It can be understood that the memory 602 can be a volatile memory or a non-volatile memory, or can include both volatile and non-volatile memories. Among them, the non-volatile memory can be a read-only memory (ROM), a programmable read-only memory (PROM), an erasable programmable read-only memory (EPROM), an electrically erasable programmable read-only memory (EEPROM), a ferromagnetic random access memory (FRAM), a flash memory, a magnetic surface memory, an optical disc, or a compact disc read-only memory (CD-ROM); the magnetic surface memory can be a disk memory or a tape memory. The volatile memory can be a random access memory (RAM), which is used as an external cache. By way of example but not limitation, many forms of RAM are available, such as a static random access memory (SRAM), a synchronous static random access memory (SSRAM), a dynamic random access memory (DRAM), a synchronous dynamic random access memory (SDRAM), a double data rate synchronous dynamic random access memory (DDR SDRAM), an enhanced synchronous dynamic random access memory (ESDRAM), a sync link dynamic random access memory (SLDRAM), a direct rambus random access memory (DRRAM).The memory 602 described in the embodiments of the present application is intended to include, but not limited to, these and any other suitable types of memories.

[0142] The methods disclosed in the embodiments of the present application above can be applied to the processor 601 or implemented by the processor 601. The processor 601 may be an integrated circuit chip with signal processing capabilities. In the implementation process, the steps of the above methods can be completed by the integrated logic circuit in the hardware of the processor 601 or instructions in the form of software. The above-mentioned processor 601 may be a general-purpose processor, a digital signal processor (DSP), or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The processor 601 can implement or execute the various methods, steps, and logic block diagrams disclosed in the embodiments of the present application. The general-purpose processor may be a microprocessor or any conventional processor, etc. Combining the steps of the methods disclosed in the embodiments of the present application, it can be directly embodied as being executed and completed by a hardware decoding processor, or executed and completed by a combination of hardware and software modules in the decoding processor. The software module may be located in the storage medium, which is located in the memory 602. The processor 601 reads the information in the memory 602 and combines its hardware to complete the steps of the foregoing methods.

[0143] In an exemplary embodiment, the device may be implemented by one or more application-specific integrated circuits (ASICs), DSPs, programmable logic devices (PLDs), complex programmable logic devices (CPLDs), field-programmable gate arrays (FPGAs), general-purpose processors, controllers, microcontroller units (MCUs), microprocessors, or other electronic components for performing the foregoing methods.

[0144] It should be understood that the "one embodiment" or "an embodiment" mentioned throughout the specification means that the specific features, structures or characteristics related to the embodiment are included in at least one embodiment of the present application. Therefore, the "in one embodiment" or "in an embodiment" that appears throughout the specification does not necessarily refer to the same embodiment. In addition, these specific features, structures or characteristics can be combined in one or more embodiments in any suitable manner. It should be understood that in various embodiments of the present application, the magnitude of the serial numbers of the above processes does not mean the sequence of execution, and the execution sequence of each process should be determined by its function and internal logic, and should not constitute any limitation to the implementation process of the embodiments of the present application. The serial numbers of the embodiments of the present application above are only for description and do not represent the advantages or disadvantages of the embodiments.

[0145] It should be noted that in this text, the term "comprising", "including" or any other variant thereof is intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements not only includes those elements, but also includes other elements not expressly listed, or further includes elements inherent to such process, method, article or device. Without further limitation, an element defined by the phrase "comprising a..." does not exclude the existence of additional identical elements in the process, method, article or device including the element.

[0146] The methods disclosed in several method embodiments provided by the present application can be arbitrarily combined without conflict to obtain new method embodiments.

[0147] The features disclosed in several product embodiments provided by the present application can be arbitrarily combined without conflict to obtain new product embodiments.

[0148] The features disclosed in several method or device embodiments provided by the present application can be arbitrarily combined without conflict to obtain new method embodiments or device embodiments.

[0149] As mentioned above, it is only the implementation manner of the present application, but the protection scope of the present application is not limited thereto. Any person skilled in the art within the technical scope disclosed by the present application can easily think of changes or substitutions, which should be covered within the protection scope of the present application. Therefore, the protection scope of the present application should be subject to the protection scope of the claimed rights.

Claims

1. A satellite positioning trajectory processing method, characterized in that: The method comprises: Acquire first trajectory data of satellite positioning reported by at least one electronic device; Preprocessing the first trajectory data to obtain preprocessed second trajectory data; Performing classification prediction on the second trajectory data to obtain at least one classification result corresponding to the second trajectory data; In the case that there is at least one abnormal trajectory classification in the at least one classification result, trajectory correction processing is performed on each of the abnormal trajectory classifications to obtain a corrected target trajectory.

2. The method according to claim 1, characterized in that The preprocessing of the first trajectory data to obtain preprocessed second trajectory data includes: Filtering the first trajectory data to obtain target data corresponding to the first trajectory data; The second trajectory data is determined based on the attribute parameters of the target data.

3. The method according to claim 2, characterized in that The attribute parameters at least include a longitude parameter, a latitude parameter, a speed parameter, and an azimuth parameter; and determining the second trajectory data based on the attribute parameters of the target data includes: The second trajectory data is determined by sequentially combining the longitude parameter, the latitude parameter, the speed parameter and the azimuth parameter.

4. The method according to claim 1, characterized in that: The performing classification prediction on the second trajectory data to obtain at least one classification result corresponding to the second trajectory data includes: The second trajectory data is input into a preset classification model to obtain the at least one classification result corresponding to the second trajectory data.

5. The method according to claim 4, characterized in that The method further comprises: Acquire initial trajectory data of satellite positioning reported by at least one of the electronic devices; The initial trajectory data is input into an initial classification model for training to obtain the preset classification model.

6. The method according to claim 1, characterized in that The abnormal trajectory classification includes a first abnormal trajectory classification; the first abnormal trajectory classification indicates that there are at least two abnormal points around the end point of the satellite positioning trajectory; when there is at least one abnormal trajectory classification in the at least one classification result, performing trajectory correction processing on each abnormal trajectory classification to obtain a corrected target trajectory, including: When the first abnormal trajectory classification exists in the at least one classification result, obtaining a first parameter corresponding to a first abnormal point and a second parameter corresponding to a second abnormal point among the at least two abnormal points; The corrected target trajectory is obtained according to the first parameter and the second parameter.

7. The method according to claim 1, characterized in that The abnormal trajectory includes a second abnormal trajectory classification; the second abnormal trajectory classification indicates a trajectory having at least one return in the satellite positioning trajectory; and in the case where there is at least one abnormal trajectory classification in the at least one classification result, performing trajectory correction processing on each abnormal trajectory classification to obtain a corrected target trajectory, including: In the case where the second abnormal trajectory classification exists in the at least one classification result, obtaining a third parameter corresponding to a first point in the satellite positioning trajectory; the first point represents a normal point in the satellite positioning trajectory; Obtaining a fourth parameter corresponding to a second point and the first point according to the third parameter; the second point represents other points in the satellite positioning trajectory except the first point; The corrected target trajectory is obtained based on the fourth parameter.

8. The method according to claim 1, characterized in that The abnormal trajectory includes a third abnormal trajectory classification; the third abnormal trajectory classification indicates that the distance between the third point and the fourth point in the satellite positioning trajectory data is greater than a preset threshold; the third point and the fourth point are any points in the satellite positioning trajectory; When there is at least one abnormal trajectory classification in the at least one classification result, performing trajectory correction processing on each abnormal trajectory classification to obtain a corrected target trajectory includes: In the case where the third abnormal trajectory classification exists in the at least one classification result, obtaining a fifth point in the satellite positioning trajectory; the fifth point is a normal point in the satellite positioning trajectory data; The third point or the fourth point is replaced by the fifth point to obtain the corrected target trajectory.

9. A satellite positioning trajectory processing device, characterized in that: The device comprises: An acquisition unit, configured to acquire first trajectory data of satellite positioning reported by at least one electronic device; A preprocessing unit, used for preprocessing the first trajectory data to obtain preprocessed second trajectory data; a classification prediction unit, configured to perform classification prediction on the second trajectory data to obtain at least one classification result corresponding to the second trajectory data; The correction processing unit is used to perform trajectory correction processing on each abnormal trajectory classification to obtain a corrected target trajectory when there is at least one abnormal trajectory classification in the at least one classification result.

10. A satellite positioning trajectory processing device, characterized in that: include: a processor and a memory for storing a computer program capable of being executed on the processor, Wherein, when the processor is used to run the computer program, it executes the steps of the method described in any one of claims 1 to 8.

11. A storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, the steps of the method according to any one of claims 1 to 8 are implemented.

Images